[PATCH] trace-cmd record: Remove redundant check of instance in allocate_instance()

[PATCH] trace-cmd record: Remove redundant check of instance in allocate_instance()

[Steven Rostedt]

From: "Steven Rostedt (Google)" <rostedt@goodmis.org>

The error path only happens after the instance has been allocated. No need
to check if it exists again.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=217127
Reported-by: Markus Elfring <Markus.Elfring@web.de>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
---
 tracecmd/trace-record.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/tracecmd/trace-record.c b/tracecmd/trace-record.c
index b094f0699a43..41d765b593b7 100644
--- a/tracecmd/trace-record.c
+++ b/tracecmd/trace-record.c
@@ -387,11 +387,9 @@ struct buffer_instance *allocate_instance(const char *name)
 	return instance;
 
 error:
-	if (instance) {
-		free(instance->name);
-		tracefs_instance_free(instance->tracefs);
-		free(instance);
-	}
+	free(instance->name);
+	tracefs_instance_free(instance->tracefs);
+	free(instance);
 	return NULL;
 }
 
-- 
2.39.2

Re: [PATCH] trace-cmd record: Remove redundant check of instance in allocate_instance()

[Steven Rostedt]

On Fri, 2 Jun 2023 15:12:11 +0200
Markus Elfring <Markus.Elfring@web.de> wrote:

> > The error path only happens after the instance has been allocated.
> > No need to check if it exists again.
> >
> > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=217127  
> 
> Will it be more appropriate to refer to an other bug report here?

I first wasn't going to bother with it, but then I started modifying
code around it where you mention this in the above mentioned bug, and
tagged that one. I can modify it to have you original one (I'll do that
internally, not with a v2).

> 
> 2023-03-04
> Improve exception handling in allocate_instance()
> https://bugzilla.kernel.org/show_bug.cgi?id=217128
> 
> 
> Will any additional adjustments become helpful for jump targets
> in this function implementation?

No need to spend cycles on this. It's a very slow path. Optimizing it
is nothing more than an academic exercise.

-- Steve